SHIRO TAKASHIMA
4478
1'01. 80
TABLE 11 THERMODYNAMIC QUASTITIES OF L~IELECTKIC POI,AKIZATION OF HEMOGLOBIN
AH*, kcal.
AF*.
kcal.
A,$*, c:11.
2.74 6.49 -14.3 5.56 7.80 - 7.7 4.05 2.30 ,5.67 13.2 9 0 18.1 "S. Takashima, to he published in Arch. Biochem. Biophys. C . P. Smyth, "Dielectric Behavior and Structure," Chap. 4, McGraw-Hill Book Co., Inc., Kew York, N. Y., 1955. S. Glasstone, K. J. Laidler and H. Eyring, "The Theory of Rate Processes," McGraw-Hill Book Co., Inc., New York, N. Y., 1941. OzHb(crysta1) COHb(so1n.)" Waterb Ice"
,3 7
38 1'T
Fig 7 -The
x
39 103
10
temperature dependence of relaxation time of oxyhemoglobin crystal, In T os l / T .
entropy although the entropy decrease in the solution is small. The entropy change of the dielectric polarization may be considered to be the sum of the increase of the breaking of the directed valence bonds of solvent molecules and the decrease due to the orientation of the molecules in the direction of the field. But the orientation of the whole molecule is unlikely in the solid sample and the polarization may occur in such a way that i t does not break the directed hydrogen bonds of water. The marked difference between the thermodynamic quantities of the solid and solution supports our conclusion that the dielectric increments have different origins. However, i t is unlikely that the dielectric increment of the protein solid has the same mechanism as that of ice, since the heat of activation of the protein solid is much smaller than that of ice and the entropy change very differ-
[COSTRIBUTIONFROM
THE
ent. This fact eliminates the possibility of the polarization of hydrogen bonds by simple elastic displacement of protons in protein molecules and their environments. These complex phenomena can only be explained by the assumption that a t least two polarization mechanisms are involved in the dielectric behavior of protein solution and solid. Probably the presence of various polar groups and their coordination to water molecules makes the dielectric properties of protein molecules very complicated. However, information on the dielectric losses is not available for the protein solids. Thus, we lack the experimental background necessary for further analysis of the dispersion of the solid sample. It is a pleasure to thank Dr. H. Uamabe of the University of Minnesota and Dr. H. P. Schwan for their advice. Also the author is indebted to Prof. H. Tamiya of the Tokyo Cniversity for his encouragement and guidance. PHILADELPHIA, PESNA.
DEPARTMENT O F BIOPHYSICS, SCHOOL O F MEDICINE, UNIVERSITY
OF
PENNSYLVASIA]
Dielectric Properties of Albumins BY SHIRO TAKASHIMA RECEIVEDDECEMBER 23, 1957 The dielectric properties of egg and human serum albumins were studied with frozen solutions between - 5 and -60". Freezing of the solution results in a decrease of dielectric increment, a finding which indicates that the dielectric polarization of these proteins is largely due to the orientational polarization. However, these frozen solutions have small hut welldefined dielectric increments. The dielectric polarization a t low temperature was discussed in terms of the Kirkwood theory.
Introduction The experiments with hemoglobin, described in the preceding paper, were extended to egg and human serum albumins, whose dielectric properties in solution were investigated by Oncley, et a1.l The dipole moments of egg and serum albumin are about 280 and 700 debye, respectively. However, the dielectric increment of serum albumin obtained by Bayley2with dry crystals is extremely small. Our results with frozen albumin solutions are essentially the same as those obtained with oxy(1) J. L. Oncley, J. D. Ferry and J. Shack, Ann. N . Y. Acad. Sci., 40, 371 (1940); J. Ferry and J. L. Oncley, T H I S J O U R N A L , 60, 1123 (1938). (2) S. T. Bayley, Trans. Faraday Soc., 47, 5 , 509 (1951).
hemoglobin. The dipole moments were calculated and compared with the theoretical values o f Kirkwood.
Experimental Measurements were made as described in the preceding paper. Commercial egg and human serum albumins were used without purification, except that the solutions were dialyzed overnight before use. The concentration of protein was about 20-25 g. per liter. The dielectric constants of egg and serum albumin and the anomalous dispersions are shown in Figs. 1 and 2. The dielectric increments per gram are 0.11 and 0.13 for egg and serum albumin, respectively, a t -.io ; the corresponding values a t -20' are approximately 0.05 and 0.06. The _____
(3) J. Kirkwood and J. B. Schumaker, Proc. .Val. Acad. S c i . , 40, 371 (1940).
DIELECTRIC PROPERTIES OF ALBUMINS
Sept. 5 , 1958
4479
I
5
1
I
I
I
1
1
1
1
0.2 0.4 0.6 0.8 1.0 1.2, Log frequency, mc. Fig. 1.-The dielectric constant of frozen solution of egg albumin. Curves from 1 to 5 were obtained a t - 5, - 10, -30, -40 and -60", respectively. -0.2
U
- 60
-40 -20 0 Temperature, OC. Fig. 4.-The temperature dependence of the relaxation time of egg and serum albumin. Curve 1, egg albumin and curve 2, serum albumin. 1.0
4
h
8
W
I W 0
Y
7 0.5 v, I v w
-0.2
-L1 - L 0 2 0.4 0 6 0.8 1.0 1 2 Logf, mc. Fig. 2.-The dielectric constant of frozen serum albumin solution. Curves 1, 2 and 3 were obtained a t - 5 , -10 and -20", respectively. L
-
-
L
0
temperature dependences of dielectric increments of both proteins are shown in Fig. 3 . Figure 4 shows temperature dependence of the relaxation times of both albumins. In both cases the relaxation time decreased rapidly between - 5 1
4[ c
i
I
l
2 /
I
01
0
-0.4
0.4 0.8 1.2 Logf, mc. Fig. 5.-The normalized dispersion curve of serum albumin. Curve l is the Debye dispersion curve, closed circles a t -5' and open circles a t -20'. 0
the relaxation time of egg albumin reaches a constant value below -20". The normalized dispersion curves of egg albumin, shown in Fig. 5 , were compared with Debye's dispersion curve. The dispersion a t -5' fits the Debye curve markedly well, but the curve a t -20' deviates considerably from it.
I
__
- 60
-40 -20 0 Temperature, "C. Fig. 3.-The temperature dependence of the dielectric increment of egg and serum albumins. Curve 1, egg albumin and curve 2, serum albumin. and -20'. The negative temperature dependence of 7 may be due to the phase shift or to a transition of polarization mechanism. However, as can be seen from the figure,
Discussion The results observed on albumin are essentially the same as those for oxyhemoglobin which were reported in the preceding paper. The dielectric increment of serum albumin solution decreases on freezing, a change which seems to be due to the loss of the freedom of rotation of the whole molecule. The relaxation time has a similar temperature dependence. However, neither albumin shows a sharp order-disorder transition; the dielectric increment values decrease gradually on freezing and approach asymptotically to a lower plateau. This was observed markedly in the case of egg albumin. As is shown in Fig. 3, the freezing of the
4480
FRANCIS P. DWYERAND FRANCIS L. GARVAN
1701.
80
solution does not give rise to a marked decrease in the dielectric increment; on the contrary the value a t -5" is larger than that of 25". 'The dipole moments were calculated as described in the preceding paper, and llsted in Table I. I t is assumed that the axial ratio of both proteins
tion, since the orientation might accompany the breaking of the directed hydrogen bonds of icc. However, tlic temperature ciepentlence of t h c . relaxation time which is shown in Fig. 4 gives the lie'tt o f activation as about 1 hcal., which indicates th,tt orientation of the whole molecuic i n ice n u y not bc the case. TABLE I As in the case of hemoglobin, the dipole moments calculated from the data on the albumins are IHJt 'hm COWPARISOX OF THE CALCULATION AVD OBSERVFD far from the theoretical values of Kirkwood; see DIPOLEM o m v r OF EGGA U D SERUM ALBIJXIAS I. This strongly suggests that proton flucTable I' ernp , !A ba tuation is a possible mechanism ior the dielectrir "C Egg alb. berum alb Egg alb. Serum alb. - 5 polarization of these proteins. 270 360 830 820 Bayley2 who worked with dry albumin crystals -10 210 300 (sphere) (sphere) observed a very small dielectric constant, but he - 20 220 460 510 also found that the presence of excess moisture in-30 180 (ellipsoid) (ellipsoid) creased the dielectric constant and dielectric loss. -40 150 However, as mentioned in the preceding paper, it -60 120 is obvious that the polarization which was observed Soln (25") 250 700 in the present experiment cannot be attributed is 59. As can be seen from the table, the dipole merely to the polarization of sorbed water. The moment of egg albumin in the frozen state is of the polarization may be due partly to the dielectric same order of magnitude as that of the solution; polarization of the sorbed water itself, but there i q however, the dipole moment of serum albumin drops the possibility that water increases the dissociation considerably on freezing. The results on egg al- of polar groups and thus may enhance the fluctunbumin may be interpreted in two ways: (1) the tion of protons. The quantitative interpretation of the anomalous molecule has the same freedom of orientation in ice as in solution; or (2) the dielectric polarization of dispersion of the solid sample requires further study, this protein is entirely due to the sorbed water or to particularly to explain the shift of the dispersion mobile proton fluctuation in solution as well as in curve and its shape between - 20 and - 30". The author is indebted to Dr. H . 'I-amabe of the ice. The asymptotic decrease of the dipole moment below - 10" to a lower value favors explana- University of Minnesota, Department of Mathetion (1). However the dimensions of the molecule matics, and to Dr. H. P. Schwan for the hel1)ful and its shape indicate that orientation of the whole discussions. molecule in ice will require a large heat of activa- PHILADELPHIA, PEXN.4. PO81
[ C O X T R I B U r l O X FROM T H E
DEPARTMENT OF CHEMISTRY,
USIVEXSITY OF S Y D U E l ]
The Resolution of the Quinquedentate Cobalt(II1) Complexes with Ethylenediaminetetraacetic Acid BY FRANCIS P. DWYER'AND FRAKCIS L. GARVAN' RECEIVED A P R I L 21, 1958 The resolution of the quinquedentate cobalt(II1) complexes with ethylenediaminetetraacetic acid with nitro, chloro ant1 hromo groups occupying the sixth coordination position has been effected with optically active cis-dinitrobis-(ethylenediamine)-cobalt( 111) chloride. Treatment of the active bromo and chloro complexes with mercury(I1) nitrate solution or solid silver oxide caused a quantitative transformation to the sexadentate complex with complete retention of configuratioti When the optically active sexadentate complex was treated with concentrated hydrochloric acid, it was converted to t h e quinquedentate complex in which a chlorine is attached to the central metal ion, with some retention of configuration. 011 reaction with ethylenediamine, all of the complexes exchanged to give tris-( ethylenediaminekobalt( 111) ion. Only the nitro complex yielded an inactive product.
Introduction The sexadentate function of ethylenediaminetetraacetic acid, (H4Y), in the anion, (ethylenediaminetetraacetato)-cobaltate(III),2has been established by the infrared studies of Busch and (1) Department of Inorganic Chemistry, The Australian National University, Canberra, Australia. (2) I t is considered desirable with organic molecules capable of multidentate chelate function that the nomenclature applied to the metal complex should distinguish clearly between groups or atoms that are coordinated and those that are unattached, especially when the stereochemistry of the complex has been established. The name, potassium (ethy1enediaminetetraacetato)-cobaltate(II1)implies cotrrdination of all carboxyl groups; whereas potassium chloro-(ethylenediaminetriacetatoacetic acid)-cobaltate(III), implies cogrdination of the
Bailar.3 The complex ion also has been separated into the optical Quinquedentate Co(111) complexes in which one carboxyl group of the organic moiety is not attached to the metal have been obtained as the acid salts [Co (HY) XI-, (S = Br, XO2) by Schwarzenbach,6 who reported that when the sexadentate anion, [ C o ( Y ) ] - , was chlorine and three of the carboxyl groups, witb the fourth carboxyl group unattached to the octahedral sphere of the complex. (3) D . H . Busch and J. C . Bailar, THISJOURNAL, 75, 4574 (1953). (4) F. P. Dwyer, E. C . Gyarfas and D. P. hlellor, 3 . P h y s . Chcm., 69, 296 (1955). ( 5 ) F. P. Dwyer and F. L. Garvan, "Inorganic Syntheses," E . Rochow, E d . , Vol. VI, in publication. (6) G. Schwarzenbach, Hclu. Chim. A d a , 32, 839 (1949).